NAVAL POSTGRADUATE SCHOOL THESIS

NAVAL POSTGRADUATE SCHOOL MONTEREY, CALIFORNIA THESIS FEASIBILITY OF UNDERWATER FRICTION STIR WELDING OF HY-80 STEEL by William C. Stewart March 2011 Thesis Advisor: Second Reader: Terry McNelley Sarath Menon Approved for public release; distribution is unlimited

THIS PAGE INTENTIONALLY LEFT BLANK

REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instruction, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188) Washington DC 20503. 1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE March 2011 4. TITLE AND SUBTITLE Feasibility of Underwater Friction Stir Welding of HY-80 Steel 6. AUTHOR(S) William Stewart 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) Naval Postgraduate School Monterey, CA 93943-5000 9. SPONSORING /MONITORING AGENCY NAME(S) AND ADDRESS(ES) N/A 3. REPORT TYPE AND DATES COVERED Master s Thesis 5. FUNDING NUMBERS 8. PERFORMING ORGANIZATION REPORT NUMBER 10. SPONSORING/MONITORING AGENCY REPORT NUMBER 11. SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the official policy or position of the Department of Defense or the U.S. Government. IRB Protocol number N/A. 12a. DISTRIBUTION / AVAILABILITY STATEMENT Approved for public release; distribution is unlimited 12b. DISTRIBUTION CODE A 13. ABSTRACT (maximum 200 words) The purpose of this thesis is to determine the feasibility of underwater friction stir welding (FSW) of high-strength; quench and temper low carbon steels that are susceptible to hydrogen-assisted cracking (HAC). The specific benefits of underwater FSW would be weld repairs of ship and submarine control surfaces and hulls without the need for drydocking and extensive environmental control procedures. A single tool of polycrystalline cubic boron nitride (PCBN) in a Tungsten-Rhenium binder was used to conduct three bead-on-plate FSW traverses, approximately 40 inches in length on 0.25 inch HY-80 steel. The first traverse was a dry weld and the second and third traverse were wet (underwater) welds, all conducted at a combination of 400 revolutions per minute and 2 inches per minute. The wet welds were conducted for the purpose of assessing the HAC susceptibility of the process. 14. SUBJECT TERMS Friction Stir Welding, Underwater, High Strength Steel, Microstructural Properties, Hardenable Alloy Steel, Weld Repair, HY-80 Steel 15. NUMBER OF PAGES 53 16. PRICE CODE 17. SECURITY CLASSIFICATION OF REPORT Unclassified 18. SECURITY CLASSIFICATION OF THIS PAGE Unclassified 19. SECURITY CLASSIFICATION OF ABSTRACT Unclassified 20. LIMITATION OF ABSTRACT NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. 239-18 UU i

THIS PAGE INTENTIONALLY LEFT BLANK ii

Approved for public release; distribution is unlimited FEASIBILITY OF UNDERWATER FRICTION STIR WELDING OF HY-80 STEEL William Chad Stewart Lieutenant, United States Navy General Science (B.S.), United States Naval Academy, 2005 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN MECHANICAL ENGINEERING from the NAVAL POSTGRADUATE SCHOOL March 2011 Author: William Chad Stewart Approved by: Dr. Terry McNelley Thesis Advisor Dr. Sarath Menon Second Reader Dr. Knox Milsaps Chairman, Department of Mechanical Engineering and Aerospace Engineering iii

THIS PAGE INTENTIONALLY LEFT BLANK iv

ABSTRACT The purpose of this thesis is to determine the feasibility of underwater friction stir welding (FSW) of high-strength, quench and temper low carbon steels that are susceptible to hydrogen-assisted cracking (HAC). The specific benefits of underwater FSW would be weld repairs of ship and submarine control surfaces and hulls without the need for dry-docking and extensive environmental control procedures. A single tool of polycrystalline cubic boron nitride (PCBN) in a Tungsten-Rhenium binder was used to conduct three bead-on-plate FSW traverses, approximately 40 inches in length on 0.25 inch HY-80 steel. The first traverse was a dry weld and the second and third traverse were wet (underwater) welds, all conducted at a combination of 400 revolutions per minute and 2 inches per minute. The wet welds were conducted for the purpose of assessing the HAC susceptibility of the process. v

THIS PAGE INTENTIONALLY LEFT BLANK vi

TABLE OF CONTENTS I. INTRODUCTION...1 II. BACKGROUND...3 III. EXPERIMENTAL PROCEDURE...5 A. MATERIAL PROCESSING...5 B. CHEMICAL TESTING...11 C. MICROSTRUCTURE ANALYSIS...12 1. Specimen Preparation...12 2. Optical Microscope Imaging...12 3. SEM Imaging...12 D. MECHANICAL TESTING...13 1. Microhardness...13 2. Charpy V-Notch...13 a. Specimen Preparation...13 b. Specimen Testing...14 3. Tensile...14 a. Specimen Preparation...14 b. Specimen Testing...15 IV. RESULTS AND DISCUSSION...17 A. VISUAL INSPECTION...17 B. CHEMICAL ANALYSIS...17 C. OPTICAL MICROSCOPY...18 D. SEM MICROSCOPY...21 E. MECHANICAL PROPERTIES...23 1. Microhardness...23 2. Charpy V-Notch Impact Test...24 3. Tensile Testing...25 V. CONCLUSION...27 A. SUMMARY OF THIS WORK...27 B. FUTURE RESEARCH...27 APPENDIX A MICROHARDNESS GRAPHS...29 APPENDIX B ANAMET REPORT...31 LIST OF REFERENCES...33 INITIAL DISTRIBUTION LIST...35 vii

THIS PAGE INTENTIONALLY LEFT BLANK viii

LIST OF FIGURES Figure 1. FSW/P Nomenclature. After [3]...2 Figure 2. Underwater chamber with cooling coil for FSW/P. From [13]...6 Figure 3. Dry FSW second plunge and traverse. From [13]...7 Figure 4. Shallow surface defects on the tool advancing side of dry FSW. From [13]...7 Figure 5. Lack of bonding on advancing of last 10 inches of weld length. From [13]...8 Figure 6. Attempts at defect free weld runs by improving surface finish and changing FSW parameters. Advancing side defect remained but weld flash Figure 7. eliminated. From [13]...8 Polycrystalline cubic boron nitride (PCBN) CS4 tool used for FSW. From [13]...9 Figure 8. Underwater FSW in progress. From [13]...10 Figure 9. Surface appearance of first underwater FSW run. From [13]...10 Figure 10. Root surface of underwater friction stir weld. From [13]...11 Figure 11. Second underwater FSW. From [13]...11 Figure 12. Machining layout...12 Figure 13. Illustration of the Vickers hardness grid pattern. Gray line represents nominal stir zone shape....13 Figure 14. Charpy V-Notch test sample. From [15]...14 Figure 15. Tensile test sample. From [16]...14 Figure 16. Base material at 2.5x. A band-like distribution is visible....19 Figure 17. Montage of micrographs of dry FSW at 2.5x...19 Figure 18. Montage of micrographs of wet FSW at 2.5x...19 Figure 19. Micrographs of the TMAZ at 2.5x magnification: Dry (l) and Wet (r)...20 Figure 20. Micrographs of the TMAZ at 10x magnification: Dry (l) and Wet (r)...20 Figure 21. Micrograph of the stir zone at 20x magnification: Dry (l) and Wet (r)...21 Figure 22. SEM Micrographs at various magnifications: (l to r) 500x, 1000x, and 5000x...21 Figure 23. Dry FSW TMAZ SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x...22 Figure 24. Dry FSW Stir Zone SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x...22 Figure 25. Wet FSW TMAZ SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x...22 Figure 26. Wet FSW stir zone SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x...22 Figure 27. Comparison of SEM Micrographs at 5000x: (l to r) Dry TMAZ, BM, Wet TMAZ....23 Figure 28. Comparison of SEM Micrographs at 5000x: Dry SZ, BM, Wet SZ...23 Figure 29. Microhardness data for Dry and Wet FSW...24 Figure 30. Charpy V-Notch DBTT: Temperature vs. Impact Energy...25 Figure 31. Tensile test with the sample in-line with the rolled direction....26 Figure 32. Tensile test with sample transverse to the rolled direction....26 ix

THIS PAGE INTENTIONALLY LEFT BLANK x

LIST OF TABLES Table 1. Measured and required chemical composition of HY-80 steel. After [14]...17 Table 2. Comparison of HY-80 to HY-100 steel...18 xi

THIS PAGE INTENTIONALLY LEFT BLANK xii

LIST OF ACRONYMS AND ABBREVIATIONS BM Base Material CNC Computer Numerical Controlled FSP Friction Stir Processing FSW Friction Stir Welding FSW/P Friction Stir Welding and/or Processing HAC Hydrogen Assisted Cracking HAZ Heat Affected Zone IPM Inches per Minute NPS Naval Postgraduate School PCBN Polycrystalline Cubic Boron Nitride RPM Rotations per Minute SEM Scanning Electron Microscope SZ Stir Zone TMAZ Thermo-Mechanically Affected Zone USN United States Navy xiii

THIS PAGE INTENTIONALLY LEFT BLANK xiv

ACKNOWLEDGMENTS First and foremost, I would like to thank my wife. Without her utmost support, I would not have completed my work here at NPS. I need to thank my family, in particular my parents, grandparents and wife s parents. Without them, none of this would have even been possible if they had not encouraged me to work hard and persevere through hard times to enjoy the good. I would like to thank Prof McNelley and Prof Menon for their guidance on my thesis. I would not have been able to complete this without them. Also, thank you to Dr. Chanman Park and Will Young for their help in the lab work and data collection. Thank you to John Mobley and his team in the Engineering Machine Shop for their help machining the samples. Thank to you Dr. Murray Mahoney and his team at Advanced Metal Products for their work on conducting the friction stir welding as quickly as possible. Additionally, thank you to all of my other professors at NPS for their support and dedication to my success. xv

THIS PAGE INTENTIONALLY LEFT BLANK xvi

I. INTRODUCTION High strength low alloy steels such as HY-80 and HY-100 are used in the quenched and tempered condition for several applications in the US Navy, especially in ship hulls, and extensive welding is carried out in both fabrication as well as repair. The strengthening of high-strength steels by martensitic transformation leaves these alloys susceptible to Hydrogen Assisted Cracking (HAC) [1]. As a result of this susceptibility, welding these metals requires extensive preparations including, among others, pre- and post-heating, filler electrode controls such as baking and storage, and control of moisture and hydrocarbons. Welding high-strength steels underwater is difficult and requires extensive and expensive additional preparations involving such special techniques, equipment and highly specialized training adding to the cost and time of repairs [2]. Los Angeles class submarines have had fatigue cracking on the control surfaces and the difficulty in welding high-strength steels requires drydocking the submarine to make repairs. Maintaining the necessary environmental controls in a drydock is difficult due to the seawater trapped inside of the control surfaces. Putting a submarine in drydock costs the Navy and taxpayers hundreds of thousands of dollars annually. Friction stir welding (FSW) and processing (FSP) are solid state processes used in the joining and processing of metals. Friction stir welding and processing have been used on the Littoral Combat Ship (LCS) and the processing of nickel aluminum bronze propellers used on Navy ships and submarines. Friction stir welding is accomplished by using a cylindrical, rotating tool with a shoulder and projecting pin in pressed into the surface of the abutting edges of the materials to be welded. The material is softened enough for the tool pin to plunge into the material until the shoulder contacts the surface via frictional and adiabatic heating. The tool traverses along the weld line to produce a weld through the severe plastic deformation in the stir zone. Figure 1 illustrates the basic FSW/P nomenclature. Numerous studies have been conducted on the friction stir welding effects of hardenable alloy steels but, to date, very few studies have been 1

conducted on underwater friction stir welding, except for the recent NPS thesis work by Lieutenant Norman Overfield (NPS Thesis, December 2010). Figure 1. FSW/P Nomenclature. After [3] Friction stir welding and processing were invented by The Welding Institute in Cambridge, United Kingdom in 1991. However, most advances have been accomplished in more recent years due to the limitations of materials available for the tool. FSW/P will become more economical as more advances are made in the tool, robust and portable equipment, techniques. However, friction stir welding and processing will likely not replace traditional methods for welding most steels. It will likely be used in niche applications. Since friction stir welding and processing is conducted below the melting point of the material, hydrogen solubility will be reduced and hydrogen resulting from decomposition of water during FSW/P is not expected. As a result, FSW could be used to produce defect-free welds in hardenable alloy steels underwater and be economically beneficial. 2

II. BACKGROUND As tool materials and designs advance, studies on friction stir welding and processing have steadily increased. Experiments on steels from ultralow carbon to ultrahigh carbon content [4, 5] have yielded defect free welds. Defect free welds have also been produced in specialized steels including DP980 (advanced high strength steel) [6] and SKD61 tool steel [7]. The chemical composition and tool parameters such as tool RPM, traversing IPM and normal force determine the microstructure of friction stir welded or processed material. In studies conducted, defect free welds have been produced using tool and traverse speed parameters ranging from 1000 RPM and 15 millimeters per second (35 IPM) [8] to 100 RPM and 25 millimeters per minute (1 IPM) [9]. Normal force data is incomplete; however, they tend to range from 5kN (1124 lbf) [8] to 40kN (9000 lbf) [10]. As a result of studies so far, each specific type of steel will require additionally studies to determine its own set of parameters and tolerances required to produce a defect free weld, as long as the weld is conducted within the set standards. A martensitic microstructure was produced in the stir zone and TMAZ, on a smaller scale, in most cases. In these cases, the temperature in the stir zone exceeded α 1 (temperature at which complete austenite forms) during the processing, therefore rapidly cooled, and formed martensite. One study produced martensite free welds by controlling the friction stir welding parameters which prevented the stir zone temperature from exceeding α 1 [11]. As a result of these studies, post-weld metallurgical properties can be controlled by controlling the friction stir welding or processing parameters. Therefore, these parameters can be modified to suit a wide range of applications and eliminate preand/or post-weld heat treatments. So far, studies evaluating the feasibility of FSW/P underwater are extremely limited. Underwater welding requires unique skills and equipment not normally available to a shipyard welder using conventional fusion welding techniques. This results in increased cost and time. U.S. Navy shipyards do not typically employ qualified 3

underwater welders. Also, industry standards for underwater welding of HY-80 or HY- 100 steels do not exist [12]. HY-80 and HY-100 alloys involve heat treating to produce tempered martensitic microstructures. As a result, these materials are susceptible to hydrogen assisted cracking. The potential of FSW/P to prevent HAC in hardenable steel alloys is the basis for this and future research. Additionally, FSW/P of HY-80 and similar steels could result in significant cost savings by eliminating the need for a dry dock to complete submarine repairs. This study was initiated to compare previous results obtained by LT Norman Overfield to FSW of HY-80 steel. HY-80 steel was obtained and a comparative study of 4142 steel to HY-80 steel was conducted to understand the influence of chemical composition on the resulting microstructures and mechanical properties. 4

III. EXPERIMENTAL PROCEDURE A. MATERIAL PROCESSING A steel plate was obtained from the Naval Surface Warfare Center Carderock Division, Bethesda, MD, that was 0.25 inches (6.4mm) thick and 26.5 inches (mm) wide by 43.5 inches (mm) long. Pre-welding chemical analysis was conducted by Anamet, Inc., Hayward, CA and confirmed the material conformed to the HY-80 steel specification. The plate was cut into three sections for base metal analysis and dry and wet (underwater) friction stir processing and for future research. Each plate measured approximately 26.5 inches (mm) long by 14.5 inches (mm) wide. One plate was sectioned to produce 25 Charpy V-notch samples and 5 tensile test samples each in both the rolled and transverse directions. The second plate was cut into two sections each measuring approximately 26.5 inches (mm) by 7.25 inches (mm). One section each was used for dry and wet (underwater) FSP. Initially, the plate was sand blasted to remove mill-scale. However, following the dry friction stir weld, the second plate was handground to be sure all mill-scale was removed. All FSW was performed parallel to the long dimension [13]. The FSP was conducted by Advanced Metal Products and MegaStir Technologies in Provo, Utah. Figure 2 is the underwater friction stir welding chamber. It is made from clear plexiglass to enable clear observation of the FSW/P process. A copper cooling coil is attached to a chiller to provide heat removal of the water and prevent the salt water from boiling off. The chamber is sealed to plate using silicone adhesive. The water in the chamber is saltwater (3.5% salt content). 5

Figure 2. Underwater chamber with cooling coil for FSW/P. From [13] A dry FSP run of approximately 25 inches was completed at 400 RPM and 2 IPM. A plunge load of greater than 15,000 pounds was applied and decreased to 10,000 pounds. A software error caused the tool to extract after a few inches of weld, requiring a second plunge and traverse (Figure 3) [13]. Following the second plunge, the first 12 inches of weld surface looked good. There were, however, occasional surface flaws on the advancing side that may have been caused by surface oxides or mill-scale but further study is required to determine the exact cause (Figure 4). As the weld progressed, the surface flaws became more pronounced and severe. In the final 10 inches of the weld run, severe lack of bonding on the advancing side is visible (Figure 5) [13]. As a result, the tool was moved to a parallel location and an attempt was made using different approaches to produce a defect free weld including improved surface preparation and different FSW parameters (Figure 6). The severe defects remained but weld flash was reduced with improved surface finishing. 6

Figure 3. Dry FSW second plunge and traverse. From [13] Figure 4. Shallow surface defects on the tool advancing side of dry FSW. From [13] 7

Figure 5. Lack of bonding on advancing of last 10 inches of weld length. From [13] Figure 6. Attempts at defect free weld runs by improving surface finish and changing FSW parameters. Advancing side defect remained but weld flash eliminated. From [13] The tool material was polycrystalline cubic boron nitride (pcbn) consolidated with a metallic binder identified as MS80. The tool design was a convex scroll shoulder with a step spiral pin (CS4) (Figure 7) [13]. 8

Figure 7. Polycrystalline cubic boron nitride (PCBN) CS4 tool used for FSW. From [13] Due to the size limitations of the underwater chamber, two approximately 10 inch underwater weld runs were completed for a total length of approximately 20 inches. Figure 8 shows the underwater weld setup in progress. Figure 9 shows a relatively good surface appearance for the first weld. A small amount of weld flash and surface oxidation is visible but the weld appears to be free of defects and full penetration. Figure 10 shows the weld root surface and the tool extraction site [13]. Underwater welding was carried out in water containing 3.5wy% NaCl in order to simulate the welding in seawater. 9

Figure 8. Underwater FSW in progress. From [13] Figure 9. Surface appearance of first underwater FSW run. From [13] 10

Figure 10. Root surface of underwater friction stir weld. From [13] For the second underwater FSW run, the chamber was moved to the other half of the plate and resealed. During the second FSW, a leaked occurred and all of the water drained from the chamber. The chamber was refilled with fresh tap water and the weld was finished [13]. Figure 11 shows the weld surface of the second wet FSW [13]. Figure 11. Second underwater FSW. From [13] B. CHEMICAL TESTING Anamet, Inc. performed chemical testing on the base material to verify that the material was HY-80 steel and to establish a baseline to which future chemical testing of FSW/P material can be compared. Anamet, Inc. determined that the material was in fact HY-80 steel [14] based on their chemical analysis. 11

C. MICROSTRUCTURE ANALYSIS 1. Specimen Preparation Figure 12 represents how the unprocessed samples were machined out of the plate. Samples were prepared using standard processes and final polishing was performed using a 0.5 micron Al 2 O 3 suspension. The prepared surfaces were etch using a 5% Nital etchant (5% HNO 3 95% Methanol). Figure 12. Machining layout 2. Optical Microscope Imaging An optical microscope was used to examine the specimens under various magnifications. Several locations were viewed such as BM, TMAZ (advancing side), and SZ. Low-magnification montages were developed to show the entire width of the SZ, left and right TMAZ, as well as BM on either side. 3. SEM Imaging A Ziess NEON40 SEM was used with field emission electron source operating at 15 kev to examine the specimens under various magnifications. The results from several locations are noted in Chapter IV. 12

D. MECHANICAL TESTING 1. Microhardness A HVS-1000 digital microhardness tester was used to micro-indent each specimen to establish a Vickers hardness profile in a grid pattern (Figure 13). Figure 13. Illustration of the Vickers hardness grid pattern. Gray line represents nominal stir zone shape. A test load of 9.8 N with a 15 sec pause was used with a loading and unloading rate of 20 N/min. 2. Charpy V-Notch a. Specimen Preparation Charpy V-Notch test specimens were machined from the base material plate using a band saw and finished with a CNC machine to ASTM standards [15]. Figure 14 represents the Charpy V-Notch. Twenty-five samples were made in break in the rolled direction and 25 samples were made to break in the transverse direction. 13

Figure 14. Charpy V-Notch test sample. From [15] b. Specimen Testing The Charpy V-Notch test was conducted using at varying temperatures ranging from approximately -200 C (Liquid Nitrogen) to 21.5 C (ambient room temperature). Low temperature tests were carried out at ice (0 C), iced saturated saltwater mixture (~ - 20 C) liquid nitrogen-cooled methanol (~ -55 C) and liquid nitrogen (- 196 C) 3. Tensile a. Specimen Preparation Tensile test specimens were machined from the three different sections using a band saw and finished with a CNC machine to ASTM standards [16]. Figure 15 represents the tensile test samples. Five samples were made to break in the rolled direction, and five samples were made to break in the transverse direction. Figure 15. Tensile test sample. From [16] 14

b. Specimen Testing Tensile testing was conducted using eight tensile test specimens machined from the base material plate. Four specimens were cut longitudinally with the rolled direction and four specimens were cut transverse to the rolled direction. All tensile tests were conducted using a strain rate of 2.1x10-3 per second. The results are discussed in Chapter IV. 15

THIS PAGE INTENTIONALLY LEFT BLANK 16

IV. RESULTS AND DISCUSSION A. VISUAL INSPECTION Surface defects are readily apparent in the dry FSW. As shown previously in Figures 3 7, there is significant weld flash and lack of bonding on the advancing side of the weld. The wet FSW runs appear to be defect free ad fully penetrated welds with the exception of minor weld flash. B. CHEMICAL ANALYSIS The testing done by Anamet, Inc. concluded that the base material has a similar chemical composition to HY-80 steel. Table 1 shows the measured and required chemical composition of HY-80 steel. Table 2 compares the chemical compositions of the HY-80 and the HY-100 steels and it can be seen that the two steels are nearly identical in chemical composition and thus difficult to distinguish by chemical analysis alone. SPECTROCHEMICAL ANALYSIS (Reported as WT%) REQUIREMENTS Ultra Service Steel (USS HY 80) Min Max Carbon (C) 0.167 0.18 Chromium (Cr) 1.52 1.00 1.8 Copper (Cu) 0.09 0.25 Manganese (Mn) 0.39 0.10 0.40 Molybdenum (Mo) 0.24 0.20 0.60 Nickel (Ni) 2.33 2.00 3.25 Phosphorous (P) 0.015 0.025* Silicon (Si) 0.24 0.15 0.35 Sulfur (S) 0.015 0.025* Titanium (Ti) <0.005 0.02 Vanadium (V) 0.01 0.03 *P + S = 0.045 Max Table 1. Measured and required chemical composition of HY-80 steel. After [14] 17

HY 80 HY 100 Density 7.87 g/cc 0.284 lb/in³ 7.87 g/cc 0.284 lb/in³ Tensile Strength, Yield >=552 Mpa >= 80000psi >= 689 Mpa >= 100000 psi Modulus of Elasticity 205 Gpa 29700 ksi 205 Gpa 29700 ksi Poissons Ratio 0.280 0.280 0.280 0.280 Shear Modulus 80.0 Gpa 11600 ksi 80.0 Gpa 11600 ksi Compostion (WT%) Min Max Min Max Carbon (C) 0.12 0.18 0.12 0.20 Chromium (Cr) 1.00 1.80 1.00 1.80 Copper (Cu) 0.25 0.25 Iron (Fe) 93.1 96.4 92.8 96.2 Manganese (Mn) 0.10 0.40 0.10 0.40 Molybdenum (Mo) 0.20 0.60 0.20 0.60 Nickel (Ni) 2.00 3.25 2.25 3.50 Phosphorous (P) 0.025* 0.25 Silicon (Si) 0.15 0.35 0.15 0.35 Sulfur (S) 0.025* 0.025 Titanium (Ti) 0.020 0.020 Vanadium (V) 0.030 0.030 *P + S = 0.045 Max Table 2. Comparison of HY-80 to HY-100 steel C. OPTICAL MICROSCOPY In the base material, lighter and darker bands parallel to the plane of the rolled sheet were visible (Figure 16). This indicates that the thermomechanical processing did not completely homogenized the material by removing all remnants of segregation and coring effects. The microstructure is suggestive of a ferritic pearlitic structure in lower strength steels and the final heat treatment to produce a tempered martensitic structure evidently did not remove these effects. 18

Figure 16. Base material at 2.5x. A band-like distribution is visible. The dry stir zone appears to be narrower than the wet stir zone. The wet TMAZ appears to be narrower than the dry TMAZ. This likely caused by the higher quenching rate of the wet FSW. The wet stir zone appears to be more homogenized as well. Figure 17. Montage of micrographs of dry FSW at 2.5x Figure 18. Montage of micrographs of wet FSW at 2.5x At the low magnification (Figure 19), the flow bands are more visible in the dry TMAZ. The wet TMAZ is narrower than the dry TMAZ. Additionally, the layering of ferrite and pearlite is still visible close to the TMAZ in both samples. 19

Figure 19. Micrographs of the TMAZ at 2.5x magnification: Dry (l) and Wet (r) At a higher magnification (Figure 20), the layering of pearlite and ferrite is more apparent as well as the flow bands in both samples. Figure 20. Micrographs of the TMAZ at 10x magnification: Dry (l) and Wet (r) At high magnification, the stir zones appear similar but the wet stir zone appears more homogenized than dry stir zone. When compared to the base material both samples exhibit significant homogenization. Under the optical microscope, it is difficult to tell whether the microstructure is martensitic or bainitic. This will be discussed later with the scanning electron microscope results. 20

Figure 21. Micrograph of the stir zone at 20x magnification: Dry (l) and Wet (r) There does not appear to be any tunneling defects or any other major defects in either stir zone. D. SEM MICROSCOPY Micrographs of the base material, TMAZ and stir zones, for both dry and wet FSW, were taken using the scanning electron microscope (SEM). It was clear that the microstructure developed as result of the FSW an untempered martensite. The homogenization of the stir zone compared to the base material was even more apparent with the SEM. Figure 22. SEM Micrographs at various magnifications: (l to r) 500x, 1000x, and 5000 21

Figure 23. Dry FSW TMAZ SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x. Figure 24. Dry FSW Stir Zone SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x. Figure 25. Wet FSW TMAZ SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x. Figure 26. Wet FSW stir zone SEM Micrographs at various magnifications: (l to r) 500x, 1000x, 5000x, 20,000x 22

Figure 27. Comparison of SEM Micrographs at 5000x: (l to r) Dry TMAZ, BM, Wet TMAZ. Figure 28. Comparison of SEM Micrographs at 5000x: Dry SZ, BM, Wet SZ. It is apparent from the SEM micrographs that there is significant change in the microstructure between the base and the stir zone and even the dry FSW and wet FSW. The wet stir zone and TMAZ show smaller grains than the dry stir zone and base material. E. MECHANICAL PROPERTIES The mechanical properties of the base material were evaluated by Vickers hardness, Charpy V-Notch impact test and the tensile test. The processed material was evaluated by Vickers hardness test. The Vickers hardness was taken across the stir zone at various depths from the sample surface. 1. Microhardness Vickers hardness values of the base material ranged from 224 to 272. This is likely due to inhomogeneous material as previously discussed (including porosity, inclusion and the layering of ferrite and pearlite present. 23

Figure 29. Microhardness data for Dry and Wet FSW From this data, there is significant hardening in the TMAZ and stir zone over the base material. This requires further research but most likely reflects martensitic transformation after FSW with insufficient time during cooling to produce tempering. The hardness across the wet FSW is more consistent. This may due to overlapping FSW runs. 2. Charpy V-Notch Impact Test Charpy V-Notch Impact testing was done to determine the Ductile-to-Brittle Transition Temperature (DBTT) of the base material to establish a baseline to compare the processed material. Charpy V-Notch test revealed that our DBTT is between -192 C and -54.5 C. DBTT below -50 C is desired as it will be outside of the operating temperature of U.S. Navy ships and submarines. The figure below shows the data in more detail. 24

Figure 30. Charpy V-Notch DBTT: Temperature vs. Impact Energy 3. Tensile Testing Tensile testing was conducted on the base material. The yield strength (~750MPa or ~108ksi) is higher than would be expected for the HY-80 steel (strength of ~550 MPa or ~80ksi) leading one to believe that we may have HY-100 vice HY-80. There is some abnormal data some of the tests. More research is required to determine the cause of the abnormality. Tensile testing on the processed material was not accomplished in this study. 25

Figure 31. Tensile test with the sample in-line with the rolled direction. Figure 32. Tensile test with sample transverse to the rolled direction. 26

V. CONCLUSION A. SUMMARY OF THIS WORK In this work, preliminary studies were carried out on the feasibility of underwater FSW of HY-80 steel. In this work. the underwater welding was carried out in 3.5wt% NaCl water. 1. To our knowledge. this forms the first study on feasibility of underwater FSW of HY-80 steel. The feasibility of underwater FSW on HY-80 or 100 has been demonstrated. 2. A non-non tempered martensitic microstructure was formed in the stir zone and this is reflected in a stir zone hardness that exceeds that of the base metal. 3. A defect free weld can be accomplished underwater. 4. Underwater welding of HY-80 or 100 steels in seawater is feasible. These conclusions show that underwater FSW of HY-80 warrants further research as more questions have been raised than have been answered by this work. The potential cost savings and decreased repair time for the Navy supports further research as well. Further research should focus on the non-tempered martensitic microstructure and how to limit or eliminated it altogether. B. FUTURE RESEARCH 1. The next logical step is conduct similar mechanical property studies on the processed material for comparison as well as hydrogen content analysis of the base material and processed material to ensure no hydrogen embrittlement is occurring. 2. The cause of and preventions for the non-tempered martensitic microstructure formation in the stir zone should be evaluated. If this cannot be controlled, the potential for FSW of HY-80 would be limited. 27

3. A tool test should be conducted to determine the best tool material, and FSW parameters to produce the best results while limiting tool wear. 4. Future material obtained for testing should be requested to include certificates verifying the material. 28

APPENDIX A MICROHARDNESS GRAPHS 29

THIS PAGE INTENTIONALLY LEFT BLANK 30

APPENDIX B ANAMET REPORT 31

THIS PAGE INTENTIONALLY LEFT BLANK 32

LIST OF REFERENCES [1] K. Masubuchi and D. C. Martin, "Mechanisms of Cracking in HY-80 Steel Weldments," Welding Journal, vol. 41, pp. 375S 384S, 1962. [2] J. H. Nixon, "Underwater repair technology," pp. 108, 2000. [3] R. S. Mishra, "Friction stir welding and processing," Materials Science Engineering.R, Reports, vol. 50, p. 1, 2005. [4] H. Fujii, "Friction stir welding of carbon steels," Materials Science Engineering.A, Structural Materials, vol. 429, pp. 50 57, 2006. [5] Y. S. Sato, "Microstructural evolution of ultrahigh carbon steel during friction stir welding," Scr. Mater., vol. 57, pp. 557 560, 2007. [6] T. Huang, "Microstructural evolution of DP980 steel during friction bit joining," Metallurgical and Materials Transactions A Physical Metallurgy and Materials Science, vol. 40, pp. 2994 3000, 2009. [7] Y. C. Chen, "Evaluation of microstructure and mechanical properties in friction stir processed SKD61 tool steel," Mater Charact, vol. 60, pp. 1471 1475, 2009. [8] S. H. Aldajah, "Effect of friction stir processing on the tribological performance of high carbon steel," Wear, vol. 267, pp. 350 355, 2009. [9] L. Cui, "Friction stir welding of a high carbon steel," Scr. Mater., vol. 56, pp. 637, 2007. [10] A. Ozekcin, "A microstructural study of friction stir welded joints of carbon steels," Int. J. Offshore Polar Eng., vol. 14, pp. 284, 2004. [11] Y. D. Chung, H. Fujii, R. Ueji and N. Tsuji, "Friction stir welding of high carbon steel with excellent toughness and ductility," Scr. Mater., vol. 63, pp. 223 226, 7, 2010. [12] K. M. Richmond. (2010). RE: Assistance needed on welding questions. Email correspondence from branch head for Code 250 Submarine Structural, Puget Sound Naval Shipyard. [13] M.W. Mahoney. Advance Metals Products, Inc. Private communication. February 2011. [14] E. Forman. Anamet, Inc. Private communication. December 2010. 33

[15] ASTM International. (2007). ASTM E23-07a: Standard Test Methods for Notched Bar Impact Testing of Metallic Materials. ASTM International. [16] ASTM International. (2009). ASTM E8/E8M-08: Standard Test Methods for Tension Testing of Metallic Materials. ASTM International. [17] Military Specification, MIL-E-16216. (1981). [18] American Society for Metals, Metals Park,OH (USA).HandbookComittee, Metals Handbook. Vol. 9: Metallography and Microstructure, Metals Park, OH (USA): AMS, 1985. [19] Military Specification, MIL-E-23765/2E, p. 10, 1994. [20] I. Maroef, "Hydrogen trapping in ferritic steel weld metal," International Materials Reviews, vol. 47, pp. 191 223, 2002. [21] T. J. Lienert, "Friction stir welding studies on mild steel," Welding Journal, vol. 82, pp. 1 9, 2003. 34

INITIAL DISTRIBUTION LIST 1. Defense Technical Information Center Ft. Belvoir, Virginia 2. Dudley Knox Library Naval Postgraduate School Monterey, California 3. Engineering and Technology Curricular Office Naval Postgraduate School Monterey, California 4. Professor Terry McNelley Naval Postgraduate School Monterey, California 5. Professor Sarath Menon Naval Postgraduate School Monterey, California 35